INDEX

Explanations

words related to dynamic processes or variables

oai_token-act-pair · gpt-3.5-turbo

references to "dynamic" concepts or elements

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 5-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.5.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.5.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

PlotsExplanationShow Test FieldDefault Test Text

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

chens

-0.90

zona

-0.87

ablishment

-0.86

hemat

-0.85

ador

-0.84

acular

-0.84

chen

-0.83

creen

-0.82

Ã±

-0.81

ACA

-0.78

POSITIVE LOGITS

Duo

0.87

duo

0.79

 destro

0.78

 enhancement

0.77

 equilibrium

0.75

 upkeep

0.75

 analy

0.74

 dynamics

0.73

 loader

0.72

 instability

0.71

Activations Density 0.016%

words related to dynamic processes or variables

references to "dynamic" concepts or elements

No Comments

No Known Activations